thr3ads.net - R devel - funfits (PR#137) [Mar 1999]

If this information is useful, please help other people find it:
Share via:

novo@uiuc.edu

1999-Mar-10 04:42 UTC

funfits (PR#137)

Hi,

I spotted an inconsistency between the output of funfits under R, more
specifically "nnreg", and the one obtained under S-plus. Notice the
difference
in the "Root MSE", "GVC" and "GVC cost=2". As a
check mark, I am also including
the data summaries, confirming that I was in fact using the same data set.
Here's my output from both systems:

===========================================================$ Splus -e
S-PLUS : Copyright (c) 1988, 1996 MathSoft, Inc.
S : Copyright AT&T.
Version 3.4 Release 1 for HP 9000 Series, HP-UX 9.0.5 : 1996
Working data will be in /mnt/cba-econ/novo/.Data
> nn.nlSummary of outout from neural net fit
see the component summary in the output list for
more details of the fitting process
call to nnreg :
nnreg(x = inc, y = imp, k1 = 1, k2 = 7)
  # hidden units DF model DF residuals Root MSE   GCV GCV cost=2
1              1        4          296    		293.7   87400      89810
2              2        7          293    		236.9   57460      60310
3              3       10          290    		217.8   49090      52660
4              4       13          287    		213.4   47580      52200
5              5       16          284   		215.8   49200      55250
6              6       19          281    		210.7   47400      54530
7              7       22          278    		213.1   49000      57790
> summary(inc) Min. 1st Qu. Median Mean 3rd Qu. Max.
 1022    1926   3181 3312    4637 6152> summary(imp)  Min. 1st Qu. Median Mean 3rd Qu. Max.
 348.8   933.5   1932 1827    2483 5015
===============================================================
===============================================================# using RedHat
5.1, Linux kernel 2.0.34
[anovo@gauss data]$ R

R : Copyright 1999, The R Development Core Team
Version 0.63.2  (January 12, 1999)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type    "?license" or "?licence" for distribution details.

R is a collaborative project with many contributors.
Type    "?contributors" for a list.

Type    "demo()" for some demos, "help()" for on-line help,
or
        "help.start()" for a HTML browser interface to help.
Type    "q()" to quit R.

[Previously saved workspace restored]
> summary(nn.nl)Summary of outout from neural net fit
see the component summary in the output list for
more details of the fitting process
call to nnreg :
nnreg(x = inc, y = imp, k1 = 1, k2 = 7)
  # hidden units DF model DF residuals Root MSE   GCV GCV cost=2
1              1        4          296    		292.5   86690      89080
2              2        7          293    		236.6   57330      60170
3              3       10          290    		214.8   47750      51220
4              4       13          287    		210.0   46120      50600
5              5       16          284    		208.4   45890      51530
6              6       19          281    		204.8   44780      51510
7              7       22          278    		194.0   40590      47870
> summary(inc)   Min. 1st Qu.  Median    Mean 3rd Qu.    Max.
   1022    1926    3181    3312    4637    6152> summary(imp)   Min. 1st Qu.  Median    Mean 3rd Qu.    Max.
  348.8   933.5  1932.0  1827.0  2483.0  5015.0
===============================================================
Alvaro Novo

ps. The library under S-plus was provided to me by one of the original authors,
Barbara Bailey (attached her computer directory). So, I guess we are safe on
that ground.

	--      --      --      --      --      --      --      --     
                              Alvaro A. Novo
                        2116 S. Orchard St., #304
                             Urbana, IL 61801

                              novo@uiuc.edu
                        anovo@ysidro.econ.uiuc.edu
              http://www.students.uiuc.edu/~novo/Alvaro.htm
                              217-337-4893
	 --      --      --      --      --      --      --      --

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

ripley@stats.ox.ac.uk

1999-Mar-10 06:29 UTC

head link

funfits (PR#137)

On Wed, 10 Mar 1999 novo@uiuc.edu wrote:
> Hi,
> 
> I spotted an inconsistency between the output of funfits under R, more
> specifically "nnreg", and the one obtained under S-plus. Notice
the difference
> in the "Root MSE", "GVC" and "GVC cost=2". As
a check mark, I am also including
> the data summaries, confirming that I was in fact using the same data set.
> Here's my output from both systems:
Yes, but does this problem have a unique solution?  Only rarely do neural
net fits not have multiple local optima. More specifically, on the basis of
the figures here it is the S-PLUS output that is wrong as the fits under R
are achieving lower minima. Note (I have not checked here) that often R is
working in double precision where S-PLUS uses floats (and this is Fortran
code, so this is likely).  In my own neural nets package, that did change
the answers using the same compiler and same version of S-PLUS: it is a
function of the ill-posed nature of the problem.  I also get different
answers with the same code and different compilers.

I do think that problems with a package should be taken up with the
author/porter:

auk% less DESCRIPTION 
Package: funfits
Version: 2.0-3
Author: S original by Doug Nychka <nychka@stat.ncsu.edu>
        R port by Albrecht Gebhardt <albrecht.gebhardt@uni-klu.ac.at>.

If you feel the need to file a report to R-bugs on a package, we do need an
example that we can reproduce and work on.  On the evidence so far I am
not convinced there is a bug here.

-- 
Brian D. Ripley,                  ripley@stats.ox.ac.uk
Professor of Applied Statistics,  http://www.stats.ox.ac.uk/~ripley/
University of Oxford,             Tel:  +44 1865 272861 (self)
1 South Parks Road,                     +44 1865 272860 (secr)
Oxford OX1 3TG, UK                Fax:  +44 1865 272595


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To:
r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Reasonably Related Threads

Search for more possibly parallel threads

R devel - Mar 1999 - funfits (PR#137)

funfits (PR#137)

funfits (PR#137)

Reasonably Related Threads